Doubao Seed 2.1 Turbo

ByteDance’s Doubao-Seed-2.1-Turbo chat model via OpenAI-compatible API. A cost-effective Seed 2.1 variant with a 256K context, deep thinking, and multimodal understanding for scaled production.

Doubao-Seed-2.1-Turbo balances quality and cost, available through AnyFast via an OpenAI-compatible interface. It fully upgrades coding, agent, and multimodal capabilities over the previous generation, and is built for scaled production workloads that need to handle large volumes of online calls reliably — with the same capability surface as Doubao-Seed-2.1-Pro.

Key capabilities

• OpenAI-compatible — Works as a drop-in replacement with the OpenAI SDK
• 256K context window — Handles project-scale context, with up to 256K tokens of output in a single response
• Quality/cost balance — Built for large-scale, standardized production workloads
• Deep thinking — Chain-of-thought reasoning, on by default, with adjustable reasoning_effort
• Multimodal understanding — Text, image, video, and document understanding (audio not supported)
• Function calling, structured output & context caching — Robust tool use, JSON output (beta), and prompt caching
• Streaming — Real-time token streaming via SSE

Output specifications

Property	Value
Input modality	Text, Image, Video, Document
Output modality	Text
Context window	256K tokens
Max output tokens	256K

Quick example

cURL

curl https://www.anyfast.ai/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "doubao-seed-2.1-turbo",
    "messages": [
      { "role": "user", "content": "Explain quantum entanglement in simple terms." }
    ]
  }'

Thinking mode

Doubao-Seed-2.1-Turbo reasons before answering by default. thinking.type is enabled unless you set it to disabled to skip reasoning for lightweight tasks. Use reasoning_effort to tune how long the model thinks.

Python

response = client.chat.completions.create(
    model="doubao-seed-2.1-turbo",
    messages=[
        {"role": "user", "content": "Design a REST API for a blogging platform."}
    ],
    extra_body={
        "thinking": {"type": "enabled"},
        "reasoning_effort": "high"
    }
)

print(response.choices[0].message.content)

reasoning_effort controls reasoning length (effective only when thinking is enabled) and accepts minimal, low, medium, and high. Default: high.

Parameters

Parameter	Type	Required	Description
model	string	Yes	Must be doubao-seed-2.1-turbo
messages	array	Yes	List of { role, content } objects
thinking	object	No	{ "type": "enabled" | "disabled" }. Controls deep thinking. Default: enabled
reasoning_effort	string	No	minimal, low, medium, high. Effective when thinking is enabled. Default: high
max_tokens	integer	No	Maximum tokens to generate (up to 262144)
temperature	float	No	0–2. Controls randomness. Default: 1
top_p	float	No	Nucleus sampling threshold. Default: 1
stream	boolean	No	Enable SSE streaming. Default: false
tools	array	No	Function tool definitions for tool use
response_format	object	No	{ "type": "json_object" } for structured JSON output (beta)
stop	string / array	No	Sequences that stop generation

API Reference

View the interactive API playground for Doubao Seed 2.1 Turbo.